Abstract: Managing stream flow regimes for riparian ecosystem restoration
Hydrologic regimes have been modified on most of the rivers in the semi-arid southwestern United States. Water diversions and impoundments have allowed for extensive urbanization and agricultural development but also have reduced stream flow rates and altered flood regimes. Depending on the nature and extent of the hydrologic changes, riparian vegetation along the region's rivers has variously declined in abundance, increased in abundance, or undergone change in species composition, diversity, and physiognomy (Briggs and Cornelius 1998; Shafroth et al. 2002). Along the floodplains of many rivers, there has been a compositional shift from Fremont cottonwood (Populus fremontii)-Goodding willow (Salix gooddingii) forests to shrublands of tamarisk (Tamarix ramosissima), a naturalized species that was intentionally introduced to the USA in the early 1800s (Horton 1964).
Many factors may contribute to shifts in plant species composition. A particular plant species may increase in abundance and expand its range simply due to increased abundance of local seed sources. The chances that the species will become abundant, however, increase if there have been changes in environmental resource levels or disturbance regimes that favor the species (Lonsdale 1999; Davis et al. 2000). If the 'new' species has adaptive traits that allow for high fitness in the new environment, and if corresponding fitness levels of the existing suite of species declines, the composition of the plant community will change.
The purpose of this talk is to discuss some of our research findings relating to the causes of compositional shifts from Fremont cottonwood-Goodding willow woodlands to tamarisk shrublands within the US Southwest, and to discuss some implications for ecosystem restoration. We focus on two hydrologic processes and conditions, flood flows and groundwater availability, that influence seedling establishment and long-term survivorship of these pioneer tree and shrub species. We demonstrate our points by examining two case study rivers, both located in the Sonoran Desert in Arizona.
Flood flows and seedling establishment. The general conditions that enable seedlings of Fremont cottonwood and Goodding willow seedlings to establish in abundance along alluvial rivers of the Sonoran desert are well understood (Stromberg 1997; Shafroth et all. 1998). Large-scale establishment typically occurs during 'El-nino' years, such as 1993 and 1995, when large winter floods are followed by declining water levels or by smaller flood peaks during spring. The large floods scour vegetation and mobilize sediment and thus deposit or expose the bare mineral soils needed for seedling establishment of these small-seeded species. Slow recession of the water levels during the spring season exposes the moist soil during the vernal period of seed dispersal. When these climatic flood patterns are altered by river damming and flow management, conditions for plant establishment change.
The flows on the Verde River in central Arizona are impounded behind two dams which are managed to provide drinking water and irrigation water to the Phoenix metropolitan area. However, both reservoirs have relatively small storage capacity. Although most of the small floods are captured, very large winter floods can exceed the reservoir storage capacity. In years with very wet winters, water has been released downstream with a relatively unmodified temporal pattern. This has allowed cottonwoods and willows to establish in the below-dam reach. During 1995, for example, the flow release pattern in the Verde was favorable for cottonwood and willow establishment above and below the dam, with large scouring flows to create seed beds followed by slowing declining water levels during the trees' period of early spring seed dispersal and germination (March-April). However, conditions in the below-dam reach also allowed for abundant establishment of tamarisk. In a study comparing riparian tree stem densities within cottonwood-willow patches, we found that stem densities of young tamarisk were more abundant in the below-dam reach than in the above-dam reach (Beauchamp and Stromberg, in prep). The most likely explanation lies with the pattern of flow release and, specifically, with small-scale modifications made to the early summer hydrograph. In the below-dam reach, there was an additional managed release that created a small peak flow in May and June. This small flood peak and subsequent water draw-down period coincided with the period of tamarisk seed dispersal, and presumably allowed for high numbers of seedlings to establish. Tamarisk is more reproductively opportunistic than Fremont cottonwood and Goodding willow, with viable seeds present over most of the growing season. It can therefore take advantage of flood pulses that occur in summer as well as spring. This opportunism may explain, in part, why tamarisk has become abundant in some below-dam reaches, where the flood flow pattern now deviates from the climatic pattern to which the local species have become adapted. Similar below-dam increases in tamarisk abundance have been observed on other flow-regulated rivers in Arizona, including the Bill Williams River below Alamo Dam (Shafroth et al. 2002).
Our knowledge of species biology is such that flow releases can be designed to favor establishment of cottonwoods, willows, and other vernal-seeding plant species that are adapted to the winter-spring flood pulse characteristic of hot-desert rivers of western USA (Rood et al. 2003). By doing so, opportunities for establishment of later-seeding species, including tamarisk, are reduced. Such an approach will not completely exclude the establishment of tamarisk, nor do we favor that as a restoration goal. The goal of restoring viable populations of characteristic regional species can be accomplished by focusing on restoring the ecosystem processes that will allow them to have high success.
Groundwater levels and tree survivorship. Our second case study focuses on the hydrologic regimes associated with long-term survivorship of Fremont cottonwood and Goodding willow trees. Here, we examine the San Pedro River, a free-flowing river located in southern Arizona. Gradients of depth to groundwater and stream flow duration exist over the length of the river due to differences in local geology and extent of groundwater pumpage for mining, agricultural and urban use. Our research shows that tamarisk is the dominant species at the drier sites while cottonwood-willow are most abundant (in absolute and relative terms) at the wetter sites. Tamarisk are deeper-rooted and more drought tolerant than Fremont cottonwood and Goodding willow and thus more likely to survive and thrive at drier sites (Shafroth et al. 2000; Horton et al. 2001). At the wetter end of the moisture gradient, the low abundance of tamarisk may be due to competitive effects of cottonwoods on tamarisk (Sher et al. 2002).
Along the San Pedro River, depth to groundwater and stream flow duration thresholds at which community dominance shifted from tamarisk to cottonwood-willow are evident. Cottonwoods and willows gave way to tamarisk as groundwater levels averaged greater than about 3 m below the floodplain surface and as stream flow duration dropped below about 75% (Lite and Stromberg 2003; and Lite and Stromberg in prep). These threshold values for groundwater depths are consistent with those reported for other rivers in the deserts of the US Southwest. Dry riverbeds where water tables are only seasonally high or where mean depths exceed 5 m may support low-density Fremont cottonwood forests. However, dense, multi-aged cottonwood-willow forests develop only along perennial or intermittent rivers where depth to ground water remains less than 3 m or 4 m (depending on the environmental context).
These findings have implications for the nature of changes in water management needed to restore cottonwood-willow to the many tamarisk-dominated rivers in southwestern USA. Politically, it can be very difficult to restore adequate stream and groundwater flows to rivers, given regional water scarcity and high demands on water for municipal, agricultural, and industrial uses. However, there are cases where re-watering is being implemented and tested as a potentially cost-effective technique to restore cottonwoods, willows, and many other hydrophytic plant species to river reaches that have become dominated by tamarisk and other drought-tolerant species (Haney 2002). And, certainly, factors other than hydrologic conditions influence the abundance of these species. Grazing by livestock or other ungulates, for example, can favor the less palatable tamarisk over the more palatable cottonwoods and willows, as can high soil salinity levels. Thus, restoration requires a holistic approach of addressing and managing all of the environmental factors that are influencing the desired end-points in terms of ecosystem structure and function.
Along some rivers, the environmental conditions favorable to cottonwood-willow forests may not realistically be restorable. For example, adequate stream flows and groundwater levels may not be available for river restoration, large flood pulses may not be desired or feasible to implement, or land managers may desire to continue to graze livestock. In such cases, we suggest being realistic about assessing the functions and values of the existing vegetation before embarking on restoration projects that involve clearing of species, such as tamarisk, that are considered to be "invasive". In some settings, such as dry or grazed river reaches, tamarisk can supply valuable habitat no longer provided by the cottonwoods and willows (Stromberg 1998a and 1998b). Without addressing the underlying physical factors driving the changes in the plant community, it is possible that vegetation-clearing efforts will cause more harm than good. In some cases, the underlying factors driving the change in species composition may have been transient and no longer operating. In other cases, however, the factors may still be continuing, necessitating continuous and repeated vegetation-clearing efforts, until the root-causes of the ecosystem change are ultimately addressed.
Beauchamp, V.B. and J. C. Stromberg. In prep. Woody vegetation responses to
flow regulation on the Lower Verde River, Arizona.
Briggs, M. K. and S. Cornelius. 1998. Opportunities for ecological improvement along the lower Colorado River and delta. Wetlands 18: 513-529.
Davis, M. A., J. O. Grime and K. Thompson. 2000. Fluctuating resources in plant communities; a general theory of invasibility. Journal of Ecology 88:528-534.
Haney, J. A. 2002. Groundwater modeling and biodiversity conservation on the Lower San Pedro River. Southwest Hydrology 1:8.
Horton, J. L., T. E. Kolb and S. C. Hart. 2001. Physiological response to groundwater depth varies among species and with river flow regulation. Ecological Applications 11:1046-1059.
Horton, J. S. 1964. Notes on the introduction of deciduous tamarisk. U.S. Forest Service Research Note RM-16.
Lite, S. J. and J. C. Stromberg. 2003. Groundwater and surface water thresholds for maintaining Populus-Salix forests, San Pedro River, southern Arizona. Pages 17-28 in Conference Proceedings, Saltcedar and Water Resources in the West Symposium, San Angelo, Texas.
Lite, S. J. and J. C. Stromberg. In prep. Water availability and flood intensity influences on woody vegetation communities across lateral and longitudinal gradients along the San Pedro River, AZ.
Lonsdale, W. M. 1999. Global patterns of plant invasions and the concept of invasibility. Ecology 80:1522-1536.
Rood, S.B., C. R. Gourley, E. M. Ammon, L. G. Heki, J. R. Klotz, M. L. Morrison, D. Mosley, G. G. Scoppettone, S. Swanson and P. L.Wagner. 2003. Flows for floodplain forests: A successful riparian restoration. BioScience 53 (7): 647-656.
Shafroth, P. B, G. T. Auble, J. C. Stromberg, D. T. Patten. 1998. Establishment of woody riparian vegetation in relation to annual patterns of streamflow, Bill Williams River, Arizona. Wetlands 18:577-590.
Shafroth, P.B., J.C. Stromberg and D.T. Patten. 2000. Woody riparian vegetation response to different alluvial water-table regimes. Western North American Naturalist 60(1):66-76.
Shafroth, P. B., J. C. Stromberg and D. T. Patten. 2002. Riparian vegetation response to altered disturbance and stress regimes. Ecological Applications 12:107-123.
Sher, A.A., D. L. Marshall, and J. P. Talor. 2002. Establishment patterns of native Populus and Salix in the presence of invasive nonnative Tamarix. Ecological Applications 12(3):760-722.
Stromberg, J. C. 1997. Growth and survivorship of Fremont cottonwood, Goodding willow, and salt cedar seedlings after large floods in central Arizona. Great Basin Naturalist 57:198-208.
Stromberg, J. C. 1998a. Dynamics of Fremont cottonwood (Populus fremontii) and saltcedar (Tamarix chinensis) populations along the San Pedro River, Arizona. Journal of Arid Environments 40:133-155.
Stromberg, J. C. 1998b. Functional equivalency of saltcedar (Tamarix chinensis) and Fremont cottonwood (Populus fremontii) along a free-flowing river. Wetlands 18:675-676.
Return to the Schedule and links to the 2003 Tamarisk Symposium
Placed on the Internet: November 2, 2003 9:52 AM
Comments on this page should be addressed to Dr.
Curtis E. Swift, Area Extension Agent, Horticulture
Colorado State Extension
2775 US Hwy 50, Grand Junction, CO. 81503